| 1 |
On Mon, Nov 06, 2006 at 01:59:33PM +0100, Hemmann, Volker Armin wrote: |
| 2 |
> On Monday 06 November 2006 13:49, Jack Lloyd wrote: |
| 3 |
> > On Mon, Nov 06, 2006 at 01:05:31PM +0100, Hemmann, Volker Armin wrote: |
| 4 |
> > > lets ask the other way round. Why should it speed up anything to have |
| 5 |
> > > X>number of cores? Instead of a single thread per core, compiling |
| 6 |
> > > happily, you have two or more competing for one core and regularly kick |
| 7 |
> > > out each others data from the cache. |
| 8 |
> > |
| 9 |
> > To account for I/O wait states |
| 10 |
> |
| 11 |
> and how often does something wait for io and how often does some data is |
| 12 |
> purged from the cache, because the other make instance is activated? |
| 13 |
> |
| 14 |
> When I switched from j2 to j1, compiling did not take any longer - but the box |
| 15 |
> way much more usable. |
| 16 |
|
| 17 |
OK. <shrug> On my dual core machine, -j3 seems to be the sweet |
| 18 |
spot. Simply because something does not work for you does not mean it |
| 19 |
is going to be universally a bad idea. |
| 20 |
-- |
| 21 |
gentoo-amd64@g.o mailing list |
Archives